slapd fails to start after reboot

[Greg Petras <gpetme@gmail.com>]

Hi - 

I'm running OpenLDAP 2.2.26. I've tried this on RedHat WS 4.0 as well
as Solaris 9. I'm using 'database bdb' with BerkeleyDB 4.2.52 (with
patches recommended by Quanah). I can get the setup to properly do the
whole Kerberos/LDAP/SASL thing, but if I reboot the server slapd
doesn't come back up properly (log output below). I guess I'd have to
do a db_recover to get this working again, but isn't there a cleaner
way to ensure that the database shuts down clean on reboot? I've read
that the proper way to shutdown slapd is via 'kill -INT', but that's
what my init script already does. Does anyone know what I'm missing
here?

Thanks,

Greg

May 12 16:22:18 lab3 slapd[1784]: @(#) $OpenLDAP: slapd 2.2.26 (May  9
2005 11:21:48) $      root@dev8:/tmp/openldap-2.2.26/servers/slapd
May 12 16:22:18 lab3 slapd[1784]: bdb_db_init: Initializing BDB database
May 12 16:22:18 lab3 slapd[1785]: bdb(dc=nyfix,dc=com): shmat: id 0:
unable to attach to shared system memory region: Invalid argument
May 12 16:22:18 lab3 slapd[1785]: bdb_db_open: dbenv_open failed:
Invalid argument (22)
May 12 16:22:18 lab3 slapd[1785]: backend_startup: bi_db_open failed! (22)
May 12 16:22:18 lab3 slapd[1785]: bdb(dc=nyfix,dc=com):
DB_ENV->lock_id_free interface requires an environment configured for
the locking subsystem
May 12 16:22:18 lab3 slapd[1785]: bdb(dc=nyfix,dc=com): txn_checkpoint
interface requires an environment configured for the transaction
subsystem
May 12 16:22:18 lab3 slapd[1785]: bdb_db_destroy: txn_checkpoint
failed: Invalid argument (22)
May 12 16:22:18 lab3 slapd[1785]: slapd stopped.
May 12 16:22:18 lab3 slapd[1785]: connections_destroy: nothing to destroy.